Detailed notes of Electric Discharge Machining

1. 29
Chapter 4
ELECTRIC DISCHARGE MACHINING
4.1 INTRODUCTION
In 1970, the English scientist, Priestley, first detected the erosive effect of electrical
discharges on metals. More recently, during research (to eliminate erosive effects on
electrical contacts) the soviet scientists, Lazarenko and Lazarenko, decided to exploit the
destructive effect of an electrical discharge and develop a controlled method of metal
machining. In 1943, they announced the construction of the first spark erosion machine.
The spark generator used in 1943, known as the Lazarenko circuit, has been employed
over many years in power supplies for EDM machines and an improved form is used in
current applications.
The EDM process can be compared with the conventional cutting process, except
that in this case, a suitably shaped tool electrode, with a. precision controlled feed
movement is employed in place of the cutting tool, and the cutting energy is provided by
means of short duration electrical impulses. EDM has found ready application in the
machining of hard metals or alloys which cannot be machined easily by conventional
methods. It thus plays a major role in the machining of dies, tools, etc., made of tungsten
carbides, stellites or hard steels.
Alloys used in the aeronautics industry, for example, hastalloy, nimonic, etc., could also
be machined conveniently by this process. This process has the added advantage of being
capable of machining complicated components.
MRR and surface finish are both controlled by the spark energy. In modern EDM
equipment, the spark energy is controlled by a dc power supply. The power supply works
by pulsing on and off the current at certain frequencies (between 10 and 500 kHz). The
on-time as a percentage of the total cycle time (inverse of the frequency) is called the
duty cycle. EDM power supplies must be able to control the pulse voltage, current,
duration, duty cycle, frequency, and electrode polarity. The power supply controls the
spark energy mainly by two parameters: current on-time and discharge current. Figure

2. 30
28-24 shows the effect of current on-time and discharge current on crater size. Larger
craters are good for high MRRs. Conversely, small craters are good for finishing
operations. Therefore, generally, higher duty cycles and lower frequencies are used to
maximize MRR. Further, higher frequencies and lower discharge currents are used to
improve surface finish while reducing the MRR. Higher frequencies generally cause
increased tool wear.
Wire EDM, shown in Figure 28-25, involves the use of a continuously moving
conductive wire as the tool electrode. The tensioned wire of copper, brass, tungsten, or
molybdenum is used only once, traveling from a take-off spool to a take-up spool while
being "guided" to produce a straight narrow kerf in plates up to 75 mm thick. The wire
diameter ranges from 0.05 to 0.25 mm with positioning accuracy up to ± 0.005 mm in
machines with NC. The dielectric is usually deionized water because of its low viscosity.
This process is widely used for the manufacture of punches, dies, and stripper plates, with
modern machines capable of routinely cutting die relief, intricate openings, tight radius
contours, and corners.
EDM is applicable to all materials that are fairly good electrical conductors, including
metals, alloys and most carbides. The hard ness, toughness, or brittleness of the material
imposes no limitations. EDM provides a relatively simple method for making holes and
pockets of any desired cross section in materials that are too hard or too brittle to be
machined by most other methods. The process leaves no burrs on the edges. About 80 to
90% of the EDM work performed in the world is in the manufacture of tool and die sets
for injection molding, forging, stamping, and extrusions. The absence of almost all
mechanical forces makes it possible to EDM fragile or delicate parts without distortion.
EDM has been used in microma-chining to make feature sizes as small as 0.01 mm
In the both vertical EDM and wire cut, the actual process is similar. The work piece
and the electrodes are immersed in a dielectric fluid such as oil or deionized water. The
electrode and work piece are separated by a small gap and voltage is applied. When there
is enough voltage, the dielectric breaks down. A spark jumps the gape, striking the work
piece and vaporizing part of the material. The intense heat also malts a small portion of
the material. The current then is pulsed off, and the dielectric into the area carrying away

3. 31
most of the melted material in the form of small chips or cinders. The current is pulsed on
and off at a rapid rate, typically at frequencies of form 500 to 1000,000 pulses per
seconds. The chips absorb most of the heat produced by the sparks. The electric fluid that
flushes away these particles also dissipates the heat enabling the tool and work piece to
remain relatively cool despite the very high, localized temperatures produced by the
spark discharge. The shapes of the electrodes create a correspondingly shaped cavity in
the work piece that is always slightly larger than the dimension of the electrodes. This
over cut reflects the size of the gap and can be small as 0.001 inch with low cutting rates.
A small potion of the melted material is redeposit on the work piece. The amount
redeposit can be minimized by carefully controlling the pulsing of the current and the
flow of the dielectric. In most applications this deposited material is not a problem and
can even be beneficial. In other applications a finishing step is required to remove it.
Other than the redeposit, EDM is a completely burr free machining method.
The popularity of the EDM process is due to the following advantages:
(i) The process can be readily applied to electrically conductive materials.
Physical and metallurgical properties of the work material, such as strength, toughness,
microstructure, etc., are no barrier to its application.
(ii) During machining, the work piece is not subjected to mechanical deformation as there
is no physical contact between the tool and work. This makes the process more versatile.
As a result, slender and fragile jobs can be machined conveniently.
(iii) Although the metal removal in this case is due to thermal effects, there is no heating
in the bulk of the material.
(iv) Complicated die contours in hard materials can be produced to a high degree of
accuracy and surface finish.
(v)The overall production rate compares well with the conventional processes because it
can dispense with operations like grinding, etc.
(vi)The surface produced by EDM consists of a multitude of small craters. This may help
in oil retention and better lubrication, especially for components where lubrication is a
problem. The random distribution of the craters does not result in an appreciable

4. 32
reduction in fatigue strength of the components machined by EDM.
(vii)The process can be automated easily requiring very little attention from the machine
operator.
4.2 SPARK EROSION MACHINING PROCESSES
Electric Discharge Machining (EDM) is the removal of materials
conducting electricity by electrical discharges between two electrodes
(work piece electrode and tool electrode), a dielectric fluid being used
.in the process. The aim of the process is controlled removal of material
from the work piece.
Following figure shows a classification of the spark erosion machining processes.
4.2.1 SINKING BY EDM
In this case, the metal removal is affected by nonstationary electrical discharges which
arc separated from each other both spatially and temporarily. This process includes those
EDM operations in which the average relative speed between the tool and the work piece
is coincident with the penetration speed in the work piece. ( Fig. 4.2 to 4.5)

5. 33
Fig.4.2 Fig.4.3
Fig.4.4 Fig.4.5
4.2.2 CUTTING BY EDM
It includes those machining operations where the work piece is cut off or notched.
Figures (4.4 and 4.7) demonstrate various EDM cutting operations.
Fig. 4.6 Fig. 4.7
4.2.3 GRINDING BY EDM

6. 34
Spark erosion grinding embraces the machining processes made with an electrode
rotating around 'an axis in addition to the normal electrode feed. The forms and
arrangements for this process are shown in Fig 4.8 to 4.11.
Fig 4.8 Fig 4.9
Fig 4.10
Fig 4.11
4.3 MECHANISM OF METAL REMOVAL
Fundamentally, the electro-sparking method of metal working involves an electric
erosion effect which connotes the breakdown of electrode material accompanying any
form of electric discharge. (The discharge is usually through a gas. liquid or in some
cases solids.) A necessary condition for producing a discharge is the ionization or the
dielectric, which is, splitting up of its molecules into ions and electrons. Consider the
case of a discharge between two electrodes (tool cathode and work anode) through a
gaseous (Fig. 4.12) or liquid medium. As soon as suitable voltage is applied across the
electrodes, the potential intensity of the electric field between them builds up, until at
some predetermined value, the individual electrons break loose from the surface of the
cathode and are impelled towards the anode under the influence of field forces (Fig.

7. 35
4.12). While moving in the inter-electrode space, the electrons collide with the neutral
molecules of the dielectric, detaching electrons from them and causing ionization. At
some time or the other, the ionization becomes such that a narrow channel of continuous
conductivity is formed.
When this happens, there is a considerable flow of electrons along the channel to
the anode, resulting in 'a momentary current impulse or discharge. The liberation of
energy accompanying the discharge leads to the generation of extremely high
temperature, between 8,000 and 12,OOO°C, causing fusion or partial vaporizations of the
metal and the dielectric fluid at the point of discharge. The metal in the form of liquid
drops is dispersed into the space surrounding the electrodes by the explosive pressure of
the gaseous products in the discharge. This results in the formation of a tiny crater at the
point of discharge in the work piece.
Fig. 4.12
Comparatively less metal is eroded from the cathode (tool) as compared to the anode
work due to the following reasons:
 The momentum with which positive ions strike the cathode surface is much less
than the momentum with which the electron stream impinges on the anode
surface.

8. 36
 A compressive force is generated on the cathode surface by the spark which helps
reduce tool wear.
Most of the EDM operations are conducted with electrodes (tool and work) immersed in
a liquid dielectric, for example paraffin, and the mechanism of sparking is similar to that
described above except that the dielectric is contaminated with conductive particles.
Furthermore, the particles removed from the electrodes due to the discharge fall in the
liquid, cool down und contaminate the area around the electrodes by forming colloidal
suspensions of metal. These suspensions, along with the products of decomposition of the
liquid dielectric are drawn into the space between the electrodes during the initial part or
the discharged process and are distributed along the electric lines of force, thus forming
current carrying 'bridges'.
Discharge then occurs along one of these bridges as a result of ionization, described
earlier. Spark discharge in liquid leads to an intense ejection of anode particles into the
surrounding space, but discharge in a gas results in the particle transfer and diffusion of
detached anode particles into the surface of the cathode. Both these phenomena are used
in metal working; the first in performing dimensional working operations, for example,
drilling, die sinking and the preparation of tool, etc.; the second is employed in operations
connected with the toughening and building up of surfaces. The spark erosion process
must be visualized as a succession of spark discharges distributed over the surface to be
eroded. The spark will pass between the electrode and work piece at that particular point
at which the electric field strength in the inter space is highest. Thus, successive spark
discharges erode the entire surface. A surface produced by this method has a pitted
appearance; the size and depth of the pits are determined by the spark energy, the nature
of work piece material and the dielectric.
4.4 SPARK EROSION GENERATORS
In the EDM process, electrical energy in the form of short duration impulses is required

9. 37
to be supplied to the machining gap. For this purpose, especially designed generators are
employed. The generators for spark erosion are distinguished according to the way in
which the voltage is transformed and also on the basis of the characteristics of discharge.
The discharge may be produced in· a controlled manner by 'natural' ignition and
relaxation, or by means of a controllable switching element, for example, electronic
valve, thyristor, transistor, etc. The discharge may take place with constant or changing
polarity.
On the basis of these facts, generators for EDM can be classified into:
(a) Relaxation generators.
(b) Rotary pulse generators.
(c) Static pulse generators.
4.4.1 Relaxation Generators
The relaxation or the R-C circuit was the first to be used in EDM. The circuit (Fig. 12)
comprises a D.C. power source that charges a capacitor 'C' across a resistance 'R'. If the
condenser is initially uncharged and the D.C. supply is switched on, a heavy current will
flow into the circuit with the condenser voltage rising continuously, as shown in Fig. 13 .
The condenser voltage at the instant t can be described by the relationship
U1(t) = Us[1-e-t/rc
] ………………………………….. (Equation.4.1)
Equation.1 predicts that the condenser voltage will approach the supply voltage (Us) with
a time constant equal to RC and after t=RC, the condenser voltage will be 63 per cent of
the supply voltage (Us). A discharge, across the working gap will occur if U1(t) equals the
breakdown voltage (Ub) of the dielectric within the gap. After the discharge, the dielectric
deionizes, the capacitor is recharged and the cycle repeats itself. The time taken to
recharge the capacitor to the breakdown voltage must be sufficient to allow the dielectric

10. 38
to deionize.
In practice, the spark gap is adjusted so that the discharge takes place
corresponding to a gap voltage of about O. 72 Us. Although a higher gap voltage would
liberate much more energy, the time required to recharge the condenser increases. Thus,
the benefit from higher energy content per spark is more than offset by a reduction in the
number of condenser discharge per unit time. It has been found in an R-C circuit, that for
a given condenser and breakdown voltage, there exists a 'certain value of 'R' that will
ensure correct length of the charging cycle.
Fig.4.13
relaxation generator, the spark repetition rate, for a given supply voltage and
capacitance, cannot be increased beyond a critical value and is determined by the speed at
which the spark gap is deionized and cleared of the debris after each discharge. Forced
circulation of the dielectric through the gap is necessary if high metal removal rates are
desired. As the working gap is of the order of 0.025-0.050 mm, forced circulation is
difficult, especially when large electrodes are involved. In such cases, a lower erosion
rate must be accepted than is possible with small size electrodes. The fundamental
advantages of relaxation circuits are their comparative cheapness, simplicity of design,

11. 39
robustness and relatively extensive range of discharge. They remain the only practical
means of generating low energy ranges and high frequencies required for fine finishing
and delicate operations. In spite of many modifications of relaxation circuits, they are
liable to result in high tool wear and slow metal removal rates, compared with other types
of generators. Moreover, interdependence of parameters, such as discharge intensity and
duration and energy values, creates a certain degree of inflexibility.
(i) Electrode Feed Control
Since, during operation, both the work piece and electrode are eroded, the feed
control must maintain a movement of the electrode towards the work piece at such a
speed that the working gap and hence, the sparking voltage remains unaltered. Since the
gap width is so small, any tendency of the control mechanism to hunt is highly
undesirable. Rapid response of the mechanism is essential and this implies a low inertia
drive. Overshooting may completely close the gap and cause a short circuit; hence, it is
essential to have rapid reversing speed with no backlash. Actuation of the control drive is
derived from an error indication signal obtained from an electrical sensing device
responsive to either the gap voltage or t e working current or both. Servo-mechanisms
affecting the movement of the electrode may be either electric-motor-driven, solenoid
operated or' hydraulically operated or a combination of these. An electric-motor-driven
type of gap controlled mechanism is shown in Fig. 14. Here, the electrode is carried in a
chuck fixed to a spindle, to which a rack is attached. The axial movement of the spindle
is controlled through a reduction gear box driven by D.C. shunt motor, which is
reversible so that the electrode can be withdrawn, should the gap be bridged by swarf or
the control mechanism cause the electrode to overshoot. The motor armature is
connected across a bridge network, the arms of which consist of a potential divider 'A'
connected across the D.C. supply, while the other arm consists of the ballast resistance
'B' and condenser 'C’ of the charging circuit, the later arms also being connected across the
supply. The control gear works as follows. Assume the electrode to be initially widely
spaced from the work piece and the current supply switched on to the condenser. This
will cause the condenser to be charged and the voltage will rise to approach the supply
voltage. The supply voltage will, therefore, prevail across one lower arm of the bridge.

12. 40
The voltage across the other arm of the bridge will depend on the potentiometer setting,
and if this setting is midway, then the voltage across the bridge (i.e. the difference
between voltages across the two lower limbs) will be half the supply voltage. This
voltage tends to rotate the motor, causing the electrode to close the gap.
(Fig.4.14) Electrode feed control in EDM
When the electrode reaches the correct position, sparking takes place and the condenser
rapidly charges and discharges so that a saw-tooth wave-form is produced across its ter-
minals. The electrode will cease to move when the average value of this voltage equals
that prevailing across the lower limb of the potentiometer. Under this condition, the
bridge is balanced and there is no armature current. Should the electrode overshoot, the
gap width will be smaller and the average condenser voltage will fall since the condenser
will no longer be able to charge up to the specific voltage. The bridge is now unbalanced
with a reverse polarity so the motor reverses and widens the gap until the correct position
is attained. If the electrode touches the work, the condenser is short circuited; causing the
supply voltage to appear across the ballast resistance, and the electrode is lifted away
from the work piece. A similar action takes place when the gap is bridged by swarf. The
required gap width can be obtained, for a given operation, by adjusting the potentiometer
setting.
(ii) Power delivered by an R-C circuit
The relaxation circuit shown in Fig. 12 can be considered to be made up of the (i)
charging circuit, and the (ii) discharging circuit.
The voltage Ui(t) across the condenser in Fig. 12 at time t is given by
Ui(t) = Us [1-e-t/Rc
] ………………………………….. (Equation. 4.2)

13. 41
and the charging current i(t) would be equal to
i(t) = C dUi(t)/ dt ………………………………….. (Equation. 4.3)
Substituting for Us and integrating
i(t) = Us /R (e- t/RC
) ………………………………… (Equation. 4.4)
Energy per spark is given by
E = 1/2 CUb
2
……………………………….... (Equation. 4.5)
or
E = ½ C [Us(1-e-t/RC
)]2
…………………….. (Equation.4.6)
where t' = charging time of the condenser up to the breakdown voltage. The power
delivered in time t' (average) would be obtained as
Elt' = Wavg = C/2t’ [ Us{1-e-t’/RC
}]2
or maximum power delivery through the circuit
dWavg/dx = 0 ………………………………….. (Equation.4.7)
where x = t’/RC
For maximum power delivery,from above equation,it is seen that x = 1.26, Substituting
for x, we get
Ub = Us [ 1 – e-t’/RC]
Or
Ub/Us = 0.72………………………………….. (Equation. 4.8)

14. 42
Thus, it is seen that for maximum power delivery through the gap, the breakdown and
supply voltage should follow the relationship given in (Equa.4.8)
(iii) Metal removal rate using relaxation circuit
Metal removal rate in EDM, using relaxation type circuit, is proportional to the product
of frequency of charging (f) and the energy delivered per spark.
Metal removal rate is proportional to f(0.5)CUb
2
Or,
Metal removal rate = K1 f(1/2 CUb
2
)
Where K1 is the constant of proportionality.
The frequency of charging would be given by,
f=1/t’
where t’,time of charging the condenser is given by
t’= RC ln[1/(1-Ub/ Us)]
The mrr is therefore given by
MRR = (K1 /2R) Ub
2
[1/ ln{1/(1-Ub/ Us)}]….. (Equation. 4.9)
From Eq. 4.8 and Fig. 15, it can be seen that for a given circuit, the metal removal rate
will increase with decreasing R. However, R can not be made very low because, in that
case, arcing will occur instead of sparking and such a situation is detrimental to the work
surface finish. The minimum value of the resistance that will prevent arcing is known as
critical resistance.

15. 43
Resistance
Fig.15
iv) Electrical parameters in R-C circuit:
The theoretical analysis of the relaxation circuit given here indicates that the four
parameters that govern the metal removal rate are:
(i) Supply voltage (UR). and breakdown voltage (Ub)
(ii) Charging resistance (R).
(iii) Capacitance (C).
(iv) Gap setting, that is, the dielectric strength of the gap.
(v) Supply voltage
Keeping all other factors constant, an increase in breakdown voltage will result in
increased, energy per spark. Consequently, the metal removal rate will increase resulting
in bigger craters on the work surface, and hence poor surface finish.
The effect of breakdown voltage and mean current on the metal removal rate is given in
Fig. 16.

16. 44
The selection of supply voltage is a compromise between several factors, for example,
size of equipment, safety of operation, etc. The D.C. supply voltage used in EDM
machines ranges between 39 and 200 V. Apart from the voltage and current, the electrode
area for a given setting has been found to influence the rate of metal removal (Fig. 17).
Figure 18 shows the effect of current density on the removal rate of steel when machined
by brass tool. Figure 19 shows the effect of pulse energy on the metal removal rate when
steel is machined by brass electrodes in kerosene medium. The pulse energy at a constant
voltage is varied by changing the size of the capacitors used.

17. 45

18. 46
Fig. 19
(vi) Charging resistance
With constant gap setting, the cutting power available varies inversely as R and this
control is normally available to the operator. However, there is a limited value of power
that can be obtained by decreasing the value of R, since there will be a certain minimum
value of R (= Rm1n) below which arcing will occur (Fig. 15).
(vii) Capacitance
An increase in capacitance also increases the energy per spark and at the same time
reduces the spark frequency for a given gap setting. The cutting power available is,
therefore, virtually unaltered but the surface finish deteriorates. The values of the
capacitance C range between 10 and 100 microfarads. The metal removal rate varies, as
shown in Fig. 20, where the minimum value of R has been used for each capacitor value.

19. 47
Fig. 20
(vii) Gap selling
Figure 21 shows how the cutting power depends upon the relative voltage to which the
capacitor is charged. If the gap is larger than the optimum, the consequent reduction in
frequency is not compensated for by an increase in energy per spark and hence, the power
falls. At shorter gaps the power decreases because the increased frequency is not suffi-
cient to compensate for the reduction in stored energy.
In practice, it becomes increasingly difficult to endure optimum conditions of gap
setting as power is increased because
o The dielectric gets contaminated with metal particles and breakdown will
occur at lower voltage.
o increasing power, R is reduced. This helps in dielectric breakdown at lower
voltages
.

20. 48
Fig. 21
(viii) Relaxation circuit with series inductance
The presence of inductance in the relaxation circuit ensures that at the beginning of the
charging cycle the current is zero. The rate of change of voltage across the capacitor and
the gap is also zero. This voltage, tending to reform the discharge across the gap, is
initially slow to rise, and hence more time is available for deionization of the dielectric,
and higher sparking frequencies are possible. Again, if the inductance is such that the
circuit is oscillatory, a further increase in charging efficiency and power is possible. It has
been reported that the modified circuit can increase the machining efficiency by 25 to 30
per cent and machining speed can be increased by about four times that attainable with
the R·C circuit.

21. 49
The main advantage of the relaxation circuits with or without inductance is that they are
simple, cheap and rugged. Fairly high frequency pulses can be generated (up to about
10,000 Hz) at low power outputs. With the relaxation circuits, the metal removal rate
under ideal conditions is limited to 2g of steel per minute.
4.4.2 Rotary Pulse Generators
In order to increase the metal removal rates, motor generator sets have been
developed to supply the required machining power in EDM. These generators are
commonly referred to as the rotary pulse generators and produce asymmetric output
waves so that the advantage of the equivalent of a D.C. power supply can be maintained.
The basic circuit of a rotary pulse generator is given in Fig, 22. During operation, the
capacitor ‘C’ is charged through the diode ‘D’ on half cycle. On the following half
cycle, the slim of the voltage from the generator and charged capacitor is applied to the
gap.
This circuit allows a standard high frequency A.C. generator to be used to produce
unidirectional pulses. This circuit permits high metal removal rates but produces exceedingly
rough surfaces.
(i) Controlled Pulse Circuit
In the electrical circuits discussed earlier, the switching device was a primary factor
in determining the frequency and the amount of energy per discharge. One element of
control lacking in the basic circuit is the ability to cut of the current in case of a short
circuit. The method of breaking the short circuit in both relaxation and pulse generator

22. 50
circuits is to withdraw the electrode mechanically. However, this takes time and could
lead to intensive work surface damage.
The need for a faster method of stopping the current in the event of short circuits
resulted in the development of circuits with electronic tubes and transistors. These
circuits are known as controlled pulse circuits (static pulse generator) and offer the
advantage of faster rate of metal removal and low tool electrode wear.
The majority of spark erosion machines currently available employ transistorized
pulse circuits which can achieve higher metal removal rates together with a high degree
of accuracy. The use of controlled pulse generators enables wide variations in pulse
duration frequency and in the intensity of spark discharges, and employs power transistor
triodes as switching devices. These are switched by low power square wave generators
and allow independent 'on' and 'off' controls, and the whole system provides
unidirectional square-wave pulses to the electrodes. A vacuum tube circuit is shown in
Fig. 23. In this case, the resistor R in the R-C circuit is replaced by a series of vacuum
tubes connected in parallel. The electronic control circuit BB turn on the tube and the
condenser gets charged, This also enables the current flow to stop in case of a short
circuit. The vacuum tube circuits require high voltages and low currents.
(ii) Vibrating Electrode System
If the tool electrode is made to vibrate so that the gap between the tool and work

23. 51
opens and closes regularly, discharge is no longer entirely dependent upon the gap
conditions. Sparking will occur when the electrodes are closest, and deionization of the
dielectric fluid can take place when the gap is large enough to prevent sparking.
The efficiency of a vibrating electrode could be enhanced when it is synchronized
with the resonant frequency, of the charging capacitance. Figure 25 shows the circuit
diagram of the vibrating electrode system with D.C. supply. Due to low circuit
resistance, the system operates very near its natural resonance frequency, and the
capacitor voltage is virtually twice, that of the supply with a corresponding increase in
power and charging efficiency. The inductor core and vibrator are combined and the
natural frequency of the vibration of the mechanical system is kept sufficiently high as
compared to the resonant frequency of the circuit. This helps to obtain the desired
electrode movement.
The charging efficiency for this type of circuit has been claimed to be of the order
of 95 per cent resulting in lower running costs. The vibration frequency is of the order of
1,000 Hz. This system has been found to be particularly suitable for metal deposition and
for toughening and roughing operations.
Fig. 25
(iii) Dielectric Fluids
For dielectric fluids to be used in the EDM process, it is essential that they should
 Remain electrically nonconductive until the required breakdown voltage
is reached, that is, they should have high dielectric strength.

24. 52
 Breakdown electrically in the shortest possible time once the breakdown
voltage has been reached.
 Quench the spark rapidly or deionize the spark gap after the
 discharge has occurred.
 Be capable of carrying away the swarf particles in suspension, away from the
working gap
 Have a good degree of fluidity
 Be cheap and easily available.
Light hydro-carbon oils seem to adequately satisfy these requirements. The common
fluids that can be used are transformer oil, paraffin oil, kerosene, lubricating oils or
various petroleum distillate fractions. Recently, distilled water has also been in place of
dielectric fluid and this has been found to permit very high metal removal rates.The
dielectric should be filtered before re-use so that the contamination of the dielectric fluid
will not affect machining accuracy. This is usually accomplished by filtration.
(iv) Flushing
Flushing is defined as the correct circulation of dielectric between the electrodes and work
piece. Suitable flushing conditions are essential to obtain the highest machining efficiency. In
order to comprehend the importance of correct flushing in EDM, it is necessary to understand
the phenomenon that occurs in the machining gap when t1ushing is absent.
To start with, the dielectric is fresh, that is, it is free from eroded particles and carbon
residue resulting from dielectric cracking, and its insulation strength is high. With
successive discharges the dielectric gets contaminated, reducing its insulation strength,
and hence, discharge can take place easily. If the density of the particles becomes too
high at certain points within the gap, bridges are formed which lead to abnormal
discharges and damage the tool as well as work electrode. This build up of the wear
debris is eliminated by flushing. Flushing in EDM is as important as any of the electrical

25. 53
parameters and should be adjusted to give that degree of contamination in the dielectric
which yields optimum results.
Flushing in EDM can be achieved by any one of the following methods.
a) Injection Flushing
b) Suction Flushing
c) Side Flushing
d) Flushing by dielectric pumping
a) Injection Flushing: The dielectric fluid is injected continuously into the working gap
either through the work piece or toll. A hole is provided in the work piece or tool for this
purpose.
b) Suction Flushing: In this method, the fluid is sucked either through the work piece or
the tool electrode. Compared with Injection Flushing, Suction avoids taper effects due to
sparking via particles along the sides of the electrode. Suction flushing through the tool
rather than through the work piece has proved to be more efficient.
c) Side Flushing: When flushing holes can’t be drilled either in the work piece or tool
this type of flushing is employed. For the entire working area to be evenly flushed,
special precautions have to be taken for the pumping of dielectric.
d) Flushing by dielectric pumping: Flushing is obtained by using the electrode
pulsation movement. When the electrode is raised, the gap increases, resulting in clean
dielectric being sucked into mix with contaminated fluid, and as the electrode is lowered,
the particles are flushed out. This method has been found particularly suitable in deep
hole drilling.
(v) Selection of Electrode material
Four main factors determine the suitability of a material for use as an electrode.

26. 54
These are:
a) The maximum MRR
b) Wear ratio.
c) Ease with which it can be shaped or fabricated to the desired shape.
d) Cost.
From purely technical considerations, it is possible to specify a material, such as silver-
tungsten alloy as the most efficient electrode providing a high metal removal rate and
very high wear ratio, but the cost of such an electrode under most conditions is, of course
prohibitive. Generally speaking, by using a sufficient number of electrodes of material
having a low wear ratio, it is possible to produce the same accuracy of machining as with
a single electrode of material with a high wear ratio. However, it has been found that the
major controlling factors of wear ratios, metal removal rates and cutting stability are the
functions of the power supply circuit for this reason, it is impossible to provide a fixed set
of rigid rules for electrode section.
(vi) Tool Electrode Design
The tool electrode must be designed as a mirror image of the work to be produced.
However, a certain amount of clearance should be provided between the tool and work
cavity produced. The magnitude of the clearance varies with the rate of metal removal,
the materials of the tool and work. Different tools may be needed for rough and fine
machining. Tables 4.3 show the effect of operating conditions on side clearance during
boring.
S.No. Rate of cutting Finish Side clearance (mm)
1 Rapid Coarse 0.5-0.6
2 Medium Medium 0.2-0.3
3 Very slow fine 0.03-0.06

27. 55
Table 4.3 Effect of Operating condition on side clearance
(v) Surface Finish
The surface produced by the EDM process consists of a multitude of small craters
randomly distributed all over the machined face. The CLA value of the surface finish in
this case ranges between 2 and 4 μ. The quality of surface mainly depends upon the
energy per spark. If the energy content is high, deeper craters will result, leading to a
poor surface (Fig. 4.26). The surface roughness (Hcla) has also been found to be
inversely proportional to the frequency of discharge.
Assuming that each spark leads to a spherical crater formation on the work surface, the
volume of metal removed per crater will be proportional to the cube of the crater depth.
Also, it is assumed at
Hcla is proportional to ‘h’
where

28. 56
Hcla = centre line average value of the surface produced, and
h= maximum crater depth.
Also
Hcla is proportional to 1/f
Therefore
Hcla is proportional to h/f
Or
Hcla = K1h/f
Where K1 = constant of proportionality.
The volume of metal removed per discharge (V1) will be equal 10 the volume of crater
produced.
Therefore
V1 = K2h3
= K3Vo
2
C
where K2 and K3 are constants.
But
Hcla is proportional to ‘h’
Therefore
Hcla = K4 Vo
2/3
C1/3
/f
Fig. 27 shows the experimental validity of the above relation.

29. 57
(vi) Machining Accuracy
a) Taper
The holes produced by this process are usually tapered due to the presence of a
frontal spark accompanied by a side spark. An exaggerated view of the hole produced is
given in Fig. 28.

30. 58
The taper at any section of the work piece has been found to be proportional to d2
. Figure 29
shows the experimental relationship obtained when carbon
was machined, using brass tool in Kerosene as the medium.
b) Overcut
Over cut in EDM is due to side sparks and is dependent on the gap length and carter
dimensions Lazarenko has shown experimentally that over cut O can be
expressed by the relationship
O = AC1/3
+B
Where A and B are constants, the values of which depend upon the tool work pair.
Dependence of the over cut on the capacitance C is shown in Fig. 30.

31. 59
4.5 CHARACTRSTICS OF SPARK ERODEDM SURFACE
In EDM, material removal is principally due to thermal phenomenon and local
temperature in the region of 8,000 to 12,OOO°C are likely to develop. This temperature will
have an effect on the structure and the mechanical properties of machined surfaces. The
effect may or may not be significant depending upon the type of work material and the
working conditions employed.
A typical cross-section of a steel specimen after machining by the EDM process, when
examined, would normally exhibit three different regions (Fig.31).
Region I A layer of molten metal, ejected and partly redeposited.
Region 2 Recast metallic layer usually referred to as white layer. The layer has no fixed
thickness and is very hard.
Region 3 An annealed layer. Thickness of the annealed layer depends upon the energy of
discharge. It has also been found that the zone is thinner if the discharges are short with
high peak currents than if they are long with low peak currents.

32. 60
In addition to the three zones described above, sometimes tiny micro-cracks can be
observed on the material surface. This occurs particularly in the machining of tungsten
carbide or other hard materials. The size of micro-cracks has been found to depend on the
type of material and the electrical parameters, such as the pulse energy and duration.
4.6 MACHINE TOOL SELECTION
A variety of EDM machines ranging from small machines to large units are now
commercially available. The factors that have to be considered in their selection are the
(i) Number of parts to be machined.
(ii) Accuracy required.
(iii) Size of the workpiece,
(iv) Depth of the cavity, and
(v) Orientation of the cavity.
Equipment must be versatile and accurate, for tool room work where a variety of work
piece configuration is encountered, EDM machine tool design and construction is a
function of the accuracy required. In cases where the positioning accuracy need not be
held closer than 0.025 or 0.050 mm, a conventional coordinate table can be used to obtain
the position read-out from the lead screw via the hand wheel dial. For higher accuracy,
an optical read-out independent of the lead screw is desirable.

33. 61
Large sized jobs require machines with high rigidity to avoid excessive deflection. High
rigidity is also essential whilst working with large sized electrodes. The electrode holding
column must be made rigid enough to support the weight of the electrode and also to
withstand the coolant back pressure, a peculiarity of this process
4.7 APPLICATIONS
Because EDM is able to create this wide variety of difficult shapes, it has become popular
for many different applications. EDM has found widespread use not only in the
manufacture of punches and dies, but also in the mold making, aerospace applications,
making extrusion dies, and the production of small holes larger than 0.015 inch and micro
holes 0.015 inch smaller. Creating small or deep slots is another important EDM
application. However its most popular application is in the making of blanking dies.
EDM can be used in advantage in many situations, and a few examples are given below.
While none of the examples may exactly fit your type of work, they may give you idea on
how to use EDM to solve some of your more difficult machining problems. In 1973 a
punch and die set used to stamp blades for an electric knife was made of carbide, without
EDM, at a cost of $12900 in 1984, it was produced in house by wire cut EDM at a cost of
$4000. the set was made of CPM-10V and ahs stamped more than 1000,000 parts. At
three- gang punch and die took about 1000 hours to manufacture using conventional
methods and required sectionalization. Using EDM took 1/5 the time and the die was
manufactured in a single piece. Machining with a solid die is more rigid than a die made
in sections and bolted together. EDM was selected as the method to produce holes in a
stainless steel medical cannula where no burrs could be tolerated.
Spark machining is used for the manufacture of tools having complicated profiles
and for a number of other components. The decision to use the spark erosion process for
either of these broad applications is usually based on one or more of the basic
characteristics inherent in the process. Spark erosion provides an economic advantage for
making stamping tools, wire drawing and extrusion dies, header dies, forging dies,
intricate mould cavities, etc. It has been extensively used for machining of exotic

34. 62
materials used in aero-space industries, refractory metals, hard carbides, and hardenable
steels.
Delicate work pieces, such as copper parts for fitting into vacuum tubes, can be
produced by this method. The work piece in this case is too fragile to withstand the
cutting tool load during conventional machining.
Sometimes accuracy requirements dictate the use of EDM for two main reasons
a) When repetitive shapes are required, they can often be produced from an easy-to-
make male electrode.
b) When machining accuracy must be maintained after heat-treatment of the part
4.8 Future Trends
Spark machining has been in production use for approximately seventeen years and is a
well-established process for producing holes and cavities in tough materials with high
precision. There are, however, quite a number of problems still to be solved to enable the
process to be adopted on an extensive basis.
The problems are:
a) Ignition of spark discharge in a contaminated flushed dielectric.
b) Mechanism of metal erosion by spark discharge
c) Distribution of discharge energy on anode and cathode during a pulse and its
relation to the discharge, conditions width of gap and electrode material
d) Distribution of temperatures and pressures in the discharge gap for different
operating times
There arc many instances of spark erosion being applied as a production process in
addition to its more usual capacity as a jobbing process. Its use as a production process is
largely the result of multiple tooling techniques which allow several components to he
eroded simultaneously. A prominent factor of multiple tooling is the development of
multi-channel techniques. Work has been conducted with as many as 77 channels and it
is highly probable that there will be a considerable increase in the number of channels

35. 63
used.
4.9 WIRE CUT ELECTRODISCHARGE MACHINING (WEDM)
Sometimes called traveling wire EDM this is a process that is similar in configuration to
band swing except in the case of WEDM the “saw” is a wire electrode of small diameter.
Material removal is affected as a result of spark erosion as the wire electrode is fed from
a spool through the work piece. In most cases, horizontal movement of the worktable,
controlled by CNC on modern machines, determines the path of cut. However, some
WED machines move the wire horizontally to define the path of cut, leaving the part
stationary. On both types of machining configurations, the wire electrode moves
vertically over sapphire or diamond wire guides, one above and one below the work
piece. The electrode wire is used only once, then discarded because the wire looses its
form after one pass through the work piece. A steady stream of deionized water or other
fluid is used to cool the work piece and electrode wire and to flush the cut area.
4.10 SUMMARY OF EDM CHARACTERISTICS
Mechanics of metal removal melting and evaporation aided by cavitation

36. 64
Medium dielectric fluid
Tool Materials Cu, brass, Cu-W alloy
Material removal rate .1 – 10
Gap 10-125 μm
Max material removal rate 5 x 103
mm3
/m
Specific power consumption 1.8 W/mm3
/m
Critical parameters Voltage, capacitance, spark gap, dielectric
circulation, melting temperature
Material Application All conducting metals and alloys.
Shape application Blind complex cavities, Micro-holes for
nozzles, Through cutting of non-circular holes,
narrow slots
Limitations High specific energy consumption; when forced
Circulation is not possible, removal rate is quite
low; surface tends to be rough for larger
removal rates, not applicable to non-conducting
materials.